The cellular otigins of both benign and malignant prostate neoplasia remain elusive. Subtypes of benign prostatic hyperplasia (BPH) have been shown to respond differentially to androgen withdrawal, 5alpha-reductase inhibitors and alpha-adrenergic blockers. Likewise, malignant prostate cancer and its precursor, high-grade prostatic intra-epithelial neoplasia (PIN), have been shown to exhibit variable hormonal responsiveness with an unpredictable length of remission and time to recurrence. Much of the uncertainties reside in the lack of basic understanding of the origins of tumor epithelium, whether from basal/stem cells, neuroendocrine cells, luminal epithelial cells, or epithelial-mesenchymal interconversion. To gain further insight into the origin of prostate epithelium, efforts have been made to establish relevant tissue culture cell lines for comparative and correlative studies. These approaches have been faced with the criticism that passage of cells in vitro often results in genotypic and phenotypic "drift," which can no longer be used with confidence to define the nature of prostate cells from the diseased state. Moreover, normal prostatic epithelial cells, either immortalized with large T antigens or transfected with ras or myc oncoproteins, have dramatic alterations in phenotype and genotype when compared to prostatic epithelial cells that progress without such specific genetic context. Faced with these deficiencies, methods of analyzing directly clinical specimens using microdissection, high-throughput cDNA microarray and comparative genomic hybridization have been used to delineate the molecular profiling of normal and diseased tissues andcells. In the proposed study, we seek to define the molecular basis of the normal and diseased prostate gland, with special emphasis on cell-lineage relationships through profiling of genes that are associated with urogenital sinus (UGS) epithelium, stromal, ventral pad, {dorsal or ventral UGS}. The hypothesis to be tested is: Differential expression of urogenital sinus "core" genes can be found during neonatal, pubertal and adult prostate development. These marker genes should prove useful in defining the origin of prostate cells in a lobe-specific and defined relationship. Sex steroid administration could affect the developmental "fate" of prostate cells normally under rigid temporal and spatial control. The aims to be pursued in this proposal are: Aim 1 Gene expression profiles will be defined in discrete UGS components, Aim 2 Confirmation of cellular location of UGS core genes in developmental stage-specific prostate glands. Aim 3 Determine changes in the pattern of UGS ?core? genes following various epigenetic stress (estradiol+ testosterone tx., lobe-specific responses to castration). The overall objectives of this proposal are to define the origin of the normal prostate cells with specific emphasis on the primordial origin of these cells based on an understanding of fetal UGS architecture and developmental capacity.